INDEX

Explanations

phrases related to significant actions or events, often involving some kind of emotional weight

oai_token-act-pair · gpt-3.5-turbo

references to specific actions or events labeled as "acts."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 corners

-0.81

 ceilings

-0.73

 Flavoring

-0.69

 edges

-0.67

 Pavilion

-0.66

 sshd

-0.65

 Pyramid

-0.63

 Wheat

-0.63

 walls

-0.63

aways

-0.61

POSITIVE LOGITS

 luck

0.90

 sabotage

0.85

 kindness

0.80

 heroism

0.78

 omission

0.76

 aggression

0.71

 desperation

0.69

asses

0.68

OUP

0.67

 vandalism

0.67

Activations Density 0.060%

phrases related to significant actions or events, often involving some kind of emotional weight

references to specific actions or events labeled as "acts."

No Comments

No Known Activations